DESCRIPTION (Applicant's Abstract): The goals of the experiments outlined in this proposal are to measure global temporal and spatial patterns of gene expression during commitment and terminal differentiation. It has been made clear from numerous studies that cells committed to different lineages differ in the pattern of genes they express. A fundamental question concerning developmental biologists is to dissect just how different growth factors and signaling systems within stem cell populations' act to generate differentiated phenotypes. It is exceedingly clear that complex problems like embryonic development are performed by groups of molecules that function in concert. If we are to understand what is happening during commitment and lineage specification, we need to look at how all the players interact with one another. To do this we need methods to simultaneously study thousands of genes or gene products. I have set up a functional genomics program here at AECOM that consists of a working cDNA microarray facility that currently uses glass slides containing 9000 different genes for quantitative measurements of global patterns of gene expression. I propose to use this system as a tool to study mammalian cell culture systems that serve as models for differentiation into neural, glial and cardiac myocyte cells. The strategy we will employ is to first measure gene expression patterns in cells during the time course of differentiation. In subsequent experiments we will genetically modify the cells to create inducible systems to individually measure the portion of the entire program that a specific transcription factor or signaling system contributes to the overall developmental program. We propose to create and study changes in gene expression in P19 lines and ES cells that have been stably transfected with inducible alleles of known neurogenic and cardiogenic transcription factors to dissect transcriptional hierarchies leading to terminal differentiation. In this way, we hope to understand how each component of a developmental program interacts, synergizes with, and regulates the activity of other components. Bioinformatics and gene clustering tools will be critical to integrate these results in the framework of normal differentiation. We have used the embryonal carcinoma (EC) F9 cell line differentiating into parietal endoderm in response to retinoic acid (RA) plus dibutyrl cyclic AMP as our initial model system because it has been extensively studied in the past and we can compare microarray results with those obtained by more traditional methods. In addition, we recently completed a study of 16 time points from 1 hr. to 13 days of RA induced neural and glial cell differentiation in P19 cells and 10 time points from l to 10 days following DMSO induced cardiocyte differentiation in P19clone6 cells. The results of these experiments and those with F9 cells outlined in the preliminary results section of this proposal validate the experimental approach mentioned above. The approach we are using should implicate functional roles for a huge number of new genes for us and others to study in greater detail.